Hello! I have a question regarding the OCR language support for print text. https://video2.skills-academy.com/en-us/azure/cognitive-services/computer-vision/language-support#print-text These languages aren't supported at this point: Greek Hebrew Thai Vietnamese Is there a plan to support them in the future? I've found some old answer from a year ago about the plans to include Greek in Q1 2023: https://video2.skills-academy.com/en-us/answers/questions/806665/computer-vision-api-call-not-working-for-greek-ima Is it still planned? What is the current timeline? Thanks, Maciej

OCR Language support for Greek, Hebrew, Thai, Vietnamese

Maciej Szulc 5

Hello!

I have a question regarding the OCR language support for print text. https://video2.skills-academy.com/en-us/azure/cognitive-services/computer-vision/language-support#print-text These languages aren't supported at this point:

Greek
Hebrew
Thai
Vietnamese

Is there a plan to support them in the future?

I've found some old answer from a year ago about the plans to include Greek in Q1 2023: https://video2.skills-academy.com/en-us/answers/questions/806665/computer-vision-api-call-not-working-for-greek-ima

Is it still planned? What is the current timeline?

Thanks, Maciej

YutongTie-MSFT 48,001 Reputation points

2023-04-15T05:48:58.5+00:00

Hello @Maciej Szulc Thanks for reaching out to us, I will check with product team to see the roadmap.
Maciej Szulc 5 Reputation points

2023-04-26T11:59:52.2833333+00:00

@YutongTie-MSFT Have you managed to check the roadmap? 😊
Thanks!
YutongTie-MSFT 48,001 Reputation points

2023-04-27T03:47:18.1666667+00:00

Hello @Maciej Szulc Thanks for your patience, I am waiting for a confirmation of the roadmap, will let you know once I get any response. Sorry for the delay.
YutongTie-MSFT 48,001 Reputation points

2023-05-03T22:39:39.2233333+00:00

Hello @Maciej Szulc

Thanks for your patience again. Now for OCR API, we have two version, one is from Computer Vision Read API, one is From Recognizer Read API.

For Computer Vision supportability, it has been postponed after June. They made it after Form Recognizer API all GA.

For Form Recognizer, Thai, VI and Greek all have been released already for preview. The Hebrew is delayed and it’s next ETA is June

I can understand you may have been very familiar with Computer Vision, but I would invite you take a quick look at Form Recognizer since it may fit more in your scenario.

Please select the Read edition that best fits your requirements.

Images: General, in-the-wild imageslabels, street signs, and posters**Computer Vision v4.0 previewOptimized for general, non-document images with a performance-enhanced synchronous API that makes it easier to embed OCR in your user experience scenarios.Documents: Digital and scanned, including imagesbooks, articles, and reportsForm Recognizer**Optimized for text-heavy scanned and digital documents with an asynchronous API to help automate intelligent document processing at scale.About Computer Vision v3.2 GA Read

Looking for the most recent Computer Vision v3.2 GA Read? Note that all future Read OCR enhancements will be part of the two new services listed above. There will be no further updates to the Computer Vision v3.2. To continue, see the Computer Vision v3.2 GA Read overview and quickstart.

Document reference- https://video2.skills-academy.com/en-us/azure/cognitive-services/computer-vision/overview-ocr#ocr-read-editions

Thanks for your support again and I hope this helps.

Regards,

Yutong
Maciej Szulc 5 Reputation points

2023-05-10T12:44:36.4766667+00:00

Hello Yutong!

Thanks a lot for a very clear and informative answer.

I see I'll need to persuade the team to switch from the Computer Vision Read API to the Form Recognizer :)

Thanks again!

Maciej
Tal Farhan 0 Reputation points

2023-06-18T06:01:40.54+00:00

Hello,

I use From Recognizer Read for documents in the Hebrew language.

When I download the JSON file and convert it to DOCX, it exports the file fine.

But when I use the code and export the output to DOCX, the text is not recognized as Hebrew but as gibberish (by API) .

I would like to know if there is a solution to this...

Thanks.
AZ 0 Reputation points

2024-05-02T06:35:10.3166667+00:00

Hi YutongTie-MSFT,

Checking if Vietnamese and Thai languages are not supported for printed, handwriting and scanned images ?

Share via

OCR Language support for Greek, Hebrew, Thai, Vietnamese